Abstract
Scientific applications are becoming data intensive, and traditional load-balance solutions require reconsideration for scaling data and computation in various parallel systems. This chapter examines state-transition applications, which is a representative scientific application that handles grand-challenging problems (e.g., weather forecasting and ocean prediction) and relates to intensive data. We propose an adaptive workload partitioning and allocation scheme for parallelizing state-transition applications in various parallel systems. Existing schemes insufficiently balance both computation of complicated scientific algorithms and increasing volumes of scientific data simultaneously. Our solution addresses this problem by introducing a time metric to unify the workloads of computation and data. System profiles in terms of CPU and I/O speeds are considered for embracing system diversity, suggesting accurate estimation of workload. The solution consists of two major components: (1) an adaptive decomposition scheme that uses the quad-tree structure to break up workload and manage data dependency; and (2) a decentralized scheme for distributing workload across processors. Experimental results from real-world weather data demonstrate that the solution outperforms other partitioning schemes, and can be readily ported to diverse systems with satisfactory performance.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
M. Fisher, J. Nocedal, Y. Trémolet, and S. Wright, “Data assimilation in weather forecasting: a case study in pde-constrained optimization,” Optimization and Engineering, vol. 10, no. 3, pp. 409–426, 2009.
A. Robinson and P. Lermusiaux, “Overview of data assimilation,” Harvard reports in physical/interdisciplinary ocean science, vol. 62, 2000.
M. Berger and S. Bokhari, “A partitioning strategy for nonuniform problems on multiprocessors,” ToC, vol. 100, no. 5, pp. 570–580, 1987.
D. Nicol, “Rectilinear partitioning of irregular data parallel computations,” DTIC Document, Tech. Rep., 1991.
F. Manne and T. Sørevik, “Partitioning an array onto a mesh of processors,” Applied Parallel Computing Industrial Computation and Optimization, pp. 467–477, 1996.
O. Beaumont, V. Boudet, F. Rastello, and Y. Robert, “Matrix multiplication on heterogeneous platforms,” TPDS, vol. 12, no. 10, pp. 1033–1051, 2001.
E. Saule, E. O. Bas, and U. V. Catalyurek, “Partitioning spatially located computations using rectangles,” in IPDPS. IEEE, 2011.
N. Wright, S. Smallen, C. Olschanowsky, J. Hayes, and A. Snavely, “Measuring and understanding variation in benchmark performance,” in DoD High Performance Computing Modernization Program Users Group Conference (HPCMP-UGC), 2009. IEEE, 2009, pp. 438–443.
T. Zou, G. Wang, M. Salles, D. Bindel, A. Demers, J. Gehrke, and W. White, “Making time-stepped applications tick in the cloud,” in SoCC. ACM, 2011, p. 20.
G. Wang and T. Ng, “The impact of virtualization on network performance of amazon ec2 data center,” in INFOCOM. IEEE, 2010, pp. 1–9.
H. Samet, “The quadtree and related hierarchical data structures,” ACM Computing Surveys (CSUR), vol. 16, no. 2, pp. 187–260, 1984.
B. H. A. Hoekstra and R. Williams, “High-performance computing and networking.”
X. Yang, Z. Yu, M. Li, and X. Li, “Mammoth: autonomic data processing framework for scientific state-transition applications,” in Proceedings of the 2013 ACM Cloud and Autonomic Computing Conference. ACM, 2013, p. 13.
J.-R. Sack and J. Urrutia, Handbook of computational geometry. North Holland, 1999.
B. Aspvall, M. M Halldórsson, and F. Manne, “Approximations for the general block distribution of a matrix,” Theoretical computer science, vol. 262, no. 1, pp. 145–160, 2001.
M. Xue, D. Wang, J. Gao, K. Brewster, and K. Droegemeier, “The Advanced Regional Prediction System (ARPS), storm-scale numerical weather prediction and data assimilation,” Meteorology and Atmospheric Physics, vol. 82, no. 1, pp. 139–170, 2003.
R. Van der Wijngaart and P. Wong, “Nas parallel benchmarks version 2.4,” NAS technical report, NAS-02-007, Tech. Rep., 2002.
“IOR HPC Benchmark,” http://sourceforge.net/projects/ior-sio/.
H. P. F. Form, “High performance fortran language specification,” 1993.
M. Grigni and F. Manne, “On the complexity of the generalized block distribution,” Parallel Algorithms for Irregularly Structured Problems, pp. 319–326, 1996.
M. J. Berger and J. Oliger, “Adaptive mesh refinement for hyperbolic partial differential equations,” Journal of computational Physics, vol. 53, no. 3, pp. 484–512, 1984.
X. Li and M. Parashar, “Hybrid runtime management of space-time heterogeneity for parallel structured adaptive applications,” TPDS, pp. 1202–1214, 2007.
Y. Kwon, M. Balazinska, B. Howe, and J. Rolia, “Skew-resistant parallel processing of feature-extracting scientific user-defined functions,” in SoCC. ACM, 2010, pp. 75–86.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer Science+Business Media New York
About this chapter
Cite this chapter
Yang, X., Li, X. (2014). Adaptive Workload Partitioning and Allocation for Data Intensive Scientific Applications. In: Li, X., Qiu, J. (eds) Cloud Computing for Data-Intensive Applications. Springer, New York, NY. https://doi.org/10.1007/978-1-4939-1905-5_6
Download citation
DOI: https://doi.org/10.1007/978-1-4939-1905-5_6
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4939-1904-8
Online ISBN: 978-1-4939-1905-5
eBook Packages: Computer ScienceComputer Science (R0)